nevada differential response pilot project interim ... differential response pilot project-interim...
TRANSCRIPT
Nevada Differential Response Pilot Project Interim Evaluation Report
Prepared for Nevada Department of Health and Human Services
by Gary L. Siegel, Ph.D., L. Anthony Loman, Ph.D.,
Jerome Cline, MS, Christine Shannon, MSW, and Lina Sapokaite, MSW
with the assistance of Ross Battle, Vera Bonsi, Jeanene Harlick,
Marcus Loman, and Nicholas Siegel
Institute of Applied Research 103 W. Lockwood, Suite 200
St. Louis, MO 63119
November 2008
ii
Executive Summary
This is the first interim report of the evaluation being conducted of the Nevada Differential Response (DR) pilot program. DR is a more flexible, service-oriented response to child maltreatment reports that are not required by state statute or policy to have a traditional CPS investigation. The DR pilot program was implemented in February 2007 in two parts of the Las Vegas metro area in Clark County, and in early 2008 it was expanded to include Washoe and Elko counties and two additional service areas in Clark County. The evaluation being conducted by the Institute of Applied Research is multi-faceted and requires the monitoring of the program over a baseline period to determine its effects on children and their families and to learn how the program can be shaped and improved to significantly impact the broader child protection system in the state. Specific findings in this report are based on very limited data and are included only to illustrate what will be learned through the study and to show very early trends, challenges and accomplishments. The Nevada DR model is a public-private partnership involving state and county agencies and community Family Resource Centers (FRCs). The pilot project is focusing on reports of child neglect, including educational, environmental, physical, medical and improper supervision. An assessment rather than a traditional CPS investigation may be offered to families with such reports unless the family has had a substantiated report in the previous three years or has had a child made a ward of the court. State statute requires an investigation of any report that includes a child under 6 who is identified as a possible victim of abuse or neglect. Reports screened for a DR assessment are referred to the local FRC. FRCs are responsible for initiating contact with these families, conducting family assessments, providing ongoing services as needed, and determining when the case should be closed. If an FRC receives a referral it considers inappropriate for a DR assessment it returns the case to the county for a traditional investigation. Screening
• During the initial pilot period (February 2007 – September 2008), 6.6 percent of child maltreatment reports in pilot areas were referred to FRCs for a DR assessment.
• The largest percentage of reports screened for a DR response (37 percent) involved families with basic needs, followed by educational neglect (22 percent), lack of supervision (16 percent), medical neglect (9 percent), and various family problems (16 percent).
• The average age of children in DR cases was 10.2 years compared with 6.3 years in reports that were investigated.
Services
• Feedback from FRC workers on an initial (but very small) sample of cases indicates that services of some kind were provided to two out of three DR families. The others did not receive direct assistance although information about service availability was provided.
• Services most frequently provided directly by FRCs were emergency food services, assistance with utilities, rent, home repairs and other basic needs, and budgeting and financial assistance. FRCs have also provided assistance with transportation, employment services, parenting classes and home management services.
iii
• FRCs also assisted families obtain services through referrals to other agencies. Such referrals were frequently made for services such as childcare, medical/dental care, alcohol or drug treatment, domestic violence-related services, and legal services.
DR Practice
• Preliminary indications are that, with minor exceptions, the Nevada DR program is being implemented by FRCs with fidelity to the pilot model.
• FRC staff appear to have a good grasp of the DR model and its difference from traditional investigations.
• A majority of FRC staff report that they feel able to intervene effectively with the DR families with which they work. The level of satisfaction with DR tends to be higher among staff in areas with less staff turnover and greater program stability.
• The level of understanding of DR is more problematic among CPS staff; some but not all grasp the essential differences between DR and traditional investigations.
Family Response
• Initial feedback from FRC workers indicates that most families they have approached have been fairly cooperative.
• Feedback from families from the initial (and very small) sample of families has been generally positive.
• A majority of initial family respondents said they were more able to care for their children now than a year ago and were more confident in their ability to deal with the issues in their lives. Most still felt stresses, often related to their financial outlook, their current job or job prospects, or the overall well-being of their children.
Program Outcomes
• FRCs have closed 61 percent of the cases that have been referred to them for a DR-family assessment. A small number of these cases have been returned to the CPS offices from which they were referred as inappropriate for DR. Outcomes in DR cases will be matched to outcomes in similar cases that receive a traditional investigation. Outcome analysis should begin within the next quarter.
Interim Conclusions and Policy Recommendations
• If DR is a relatively minor component in the state’s child protection system it will be limited in the leverage it can exert on the system as a whole. The larger the proportion of maltreatment reports that receive a DR assessment the larger the possibility this component can impact the system overall.
• DR introduces a CPS component in Nevada that is family-centered and service oriented. But many families with very young children and chronic problems that could benefit from such an approach are excluded by current eligibility criteria.
• For the potential impact of DR to be expanded, the percentage of reports referred for a family assessment will need to be increased. To accomplish this, two things should be considered: 1) removing certain eligibility restrictions and 2) increasing the system’s capacity to provide family assessments. 1. DR Eligibility Criteria. Given current policies, the maximum percentage of cases
that can be expected to be referred for a DR assessment is about 17 percent of all reports. This level is adequate for an initial test of the DR approach, but its potential impact on the child protection system as a whole is limited, no matter how positive the outcomes achieved by the new approach. The following should be considered:
a. Remove the restriction on families with a substantiated report in the last three years, or, at least, on families with prior substantiations that involved allegations that could now be referred for a DR assessment.
iv
b. Remove the restriction on allegations involving less severe physical abuse, such as inappropriate discipline.
c. Amend state statutes to permit a DR assessment for at least some reports in which the alleged victim is under 6.
2. System Capacity. If the pool of eligible DR families is increased substantially, the capacity of the system to serve them will also need to be increased.
a. DR capacity may be increased by expanding contracts with FRCs and/or the direct involvement of CPS staff in DR.
b. Involving CPS in DR, and placing DR in direct proximity with investigations, has the added potential of increasing the service orientation of traditional investigative interventions.
c. Providing DR-related training to CPS staff would benefit the current DR program and be essential in preparing these staff for direct DR responsibilities.
• The Nevada child protection system varies across the state. The operating principle should be: What can be done in one place, should be done there, and not be postponed because it cannot be done everywhere at once.
Accomplishments
• New expertise has been developed rapidly at the state, county and community level. • Training has been provided to administrative, supervisory and field-level staff. • Collaborative procedures between state, county and community organizations have
been designed and successfully put into place. • The complex DR model has been built onto Nevada’s unique tri-level CPS structure. • DR has been implemented in the state’s most populous regions with fidelity to the
program model. • The steering committee of key representatives of the tri-level structure is an
effective instrument for reviewing policy and practice issues, addressing challenges, and guiding program implementation and modifications.
Table of Contents
Executive Summary ............................................................................................................ ii-iv Introduction .............................................................................................................................. 2 The Nevada Model ................................................................................................................... 6 Demographic Context .............................................................................................................. 9 CPS Context ........................................................................................................................... 11 Reports Screened for Differential Response .......................................................................... 16 County DR Programs ............................................................................................................. 22
1. Clark County ......................................................................................................... 22 2. Elko County ........................................................................................................... 26 3. Washoe County ..................................................................................................... 27
Survey of FRC and CPS Staffs .............................................................................................. 29
1. Survey of FRC Staffs ............................................................................................ 29 2. Survey of CPS Staffs ............................................................................................. 34
Case-Specific Survey ............................................................................................................. 39 Family Surveys ...................................................................................................................... 42 Outcome Analysis .................................................................................................................. 44
1. Comparison Group Analyses ................................................................................. 45 2. The North Carolina Family Assessment Scale ...................................................... 46 3. Preparation for Comparison Group Analyses ....................................................... 47
Interim Considerations ........................................................................................................... 49
1. Capacity, System Impact and Policy ..................................................................... 49 2. The Service Anomaly and System Adjustments ................................................... 51
Conclusion ............................................................................................................................. 53
2
Introduction
This is the first interim report of the evaluation being conducted of the Nevada Differential Response (DR) pilot program by the Institute of Applied Research (IAR). The evaluation is examining the implementation and operation of the program, its outcomes and affects on families and children, and its cost effectiveness.
The DR pilot project was implemented initially in February 2007 in two parts of Las
Vegas, the service zones of the East and South offices of Clark County Department of Family Services (CCDFS). In January 2008 the DR project was implemented in Washoe County and in February 2008 in Elko County. In March 2008 the project was expanded in Las Vegas to include the service zones of the Central and North offices of CCDFS. (Map 1 highlights the three DR program counties; Map 2 shows the Las Vegas locations in Clark County in which the DR program is now operating.) The original plan was for the Differential Response (DR) program to be phased in across the state, although the timetable is not fixed.
What is Differential Response? Differential response represents a relatively new approach to child protection. It arises from the view that it is in the best interest of children and their families that not all child maltreatment reports should be treated the same. For decades, the traditional response to child maltreatment reports has uniformly involved a formal investigation of all “accepted” reports. Accepted reports are those that meet a state’s threshold statutory requirement for a response from the child protection system (CPS). The investigation of accepted reports has historically focused on the specific allegations of child abuse or neglect, much like a report of suspected criminal behavior. As in the case of a suspected crime, a traditional investigation of a report of child maltreatment has sought to find evidence that the specific report can be substantiated and, if it can, determine what can and should be done to ensure the safety of the child. The introduction of differential response recognizes that there are significant differences among the many child maltreatment allegations that are reported, some much more serious than others, and that the response should vary in some measure that is congruent with the report.
While differential response has begun to be implemented in more states, there is one
general model that is most commonly seen. This model involves the differentiation of reports
3
CLARK
WASHOE
CARSON CITY
ELKO
DOUGLASLYON
NYE
CHURCHILL
HUMBOLDT
WHITE PINE
PERSHING
LANDER
MINERAL
LINCOLN
EUREKA
ESMERALDA
WASHOE
South Clark
East ClarkCentral Clark
North Clark
Map 1. Nevada Counties with DR Programs
Map 2. Clark County Service Areas with DR Programs
4
into two groups. The first group includes allegations of a more severe nature that may involve criminal acts and/or represent an imminent safety threat to the child. Reports in this group are judged to require a traditional investigative response, sometimes with co-investigating police authorities accompanying child protection staff. The second group of reports involves allegations of problems or situations of a less severe nature, often involving conditions that are more chronic and less acute and in which the risk to the child is real but not imminent. This second group of reports has come to be viewed as benefiting more from a broader assessment of the family situation and their living conditions and habits, from an examination of the underlying causes of current problems, from a less threatening and more friendly approach that offers support and assistance and seeks the family’s cooperation in working through issues of concern and identifying its own internal strengths and its natural support system. While the second approach, the family assessment, also focuses first on the safety of the child, its first priority is not identifying and accusing a perpetrator, but understanding and untangling the broader dynamics of the family and enlisting the help of everyone in the family in resolving and improving the situation.
Before proceeding, it should be noted that the term differential response can be
misleading. It was initially coined to refer not to a specific type of response but to a child protection system in which more than one response to child maltreatment reports was permitted.1 Missouri, one of the first states to implement a differential response approach, has a two-response system that includes traditional investigations and family assessments. Minnesota, another early implementer of differential response, initially called its new response track the “alternative response” (that is, non-investigatory), but now also uses the term family assessment for this approach. Ohio, which is currently implementing the Minnesota CPS model, refers to the new response track as the alternative response. However, as differential response has become more common (and some form has been implemented in more than 15 states at this point2
1 Child Welfare Information Gateway Issue Brief (2008). Differential Response to Reports of Child Abuse and Neglect. Washington, DC: US Department of Health and Human Services.
), it is often the case that the term differential response refers both to the multiple or dual track response system but also is the name given for the new, non-investigative response. This is the case in Nevada. Throughout this report, therefore, the term differential response will often be used to refer to the new, non-investigative response given to certain reports of child maltreatment. Where differential response has been implemented, then, accepted reports of child maltreatment receive either a 1) formal, traditional investigation (always the case for more serious allegations) or 2) a differential response, also referred to as a family assessment. When distinctions are being made between these two response types, the short-hand TR (for traditional investigation) and DR (for differential response) will also be used.
2 National Study on Differential Response in Child Welfare (2006). American Humane and Child Welfare League. Washington, DC. (retrievable at: www.americanhumane.org/site/DocServer/National_Study.pdf?ID=4761)
5
What is Different about Differential Response? Unlike a traditional investigation, a
differential response (or family assessment) does not seek to substantiate the allegations in a maltreatment report. A differential response emphasizes assistance more and apprehension less. It does not focus on the reported incident other than by way of explaining to the family what precipitated the interest of the child protection agency and as a guide to establishing the immediate safety of the child. A differential response has a wider focus on problems that may exist within the household and that may put a child’s well-being at risk. Further, a differential response involves discovery and remediation strategies different from traditional investigations.
Discovery
. A differential response is characterized by non-accusatory, non-confrontational meetings with the family. It seeks to be positive, supportive, holistic, family-centered and, overall, “family friendly.” It seeks in all cases to engage family members as a unit if possible in assessing the current state of affairs, the safety of children, the strengths of the family and the problems or issues that need to be addressed.
Remediation
. Essential to differential response is gaining the cooperation of family members from the start, identifying and building on the family’s strengths and directly involving family members in planning and decision making about what can and should be done. The objective is to help the family address its problems through its own internal resources to the extent possible, with assistance from its natural social support network, and through services that address basic needs and underlying problems that may be present and jeopardize the welfare of children.
The Evaluation. The Institute of Applied Research is conducting a multi-faceted evaluation of Nevada’s differential response pilot program. IAR has conducted similar evaluations of DR programs in other states and these states provide a frame of reference both for the systems change that differential response represents and the possibilities it has for impacting the child protection system. The design and research methodology being employed in the evaluation of the Nevada program has been shaped by what has been learned in previous research. The evaluation is scheduled to monitor and provide feedback on Nevada’s DR program for three years. This is the first interim report on the study.
Of central importance for the evaluation are regular extracts from Nevada’s child welfare
information system, UNITY, which contains current and historical data on child maltreatment reports, investigations and outcomes. These data are supplemented with case-specific surveys of workers to collect more detailed information on samples of families. These data will permit an analysis of case processing, services and outcomes. Site visits and staff interviews are being conducted of county CPS offices and Family Resource Centers to examine the manner in which
6
the new program has been implemented and to gain a full understanding of the state’s traditional child protection system. County CPS and FRC staff are also being surveyed, as are families, to obtain attitudinal and experiential feedback from those directly utilizing differential response and those most affected by it. Finally, cost data will be gathered and analyzed to determine the cost effectiveness of the differential response.
This report includes analysis of UNITY data related to the DR pilot from February 2007
through September 2008. While this report was being prepared two extracts were received. The first covered the period from February 2007 through August 2008, and data from this extract was used in most of the analysis in this report. In mid October another extract was received and was available for the discussion of the outcome analysis. Unless otherwise stipulated, references in this report to the DR pilot period means the 19 months from February 2007 through August 2008.
The Nevada Model The Nevada differential response model is a public-private partnership that involves the
state and county governments and community service organizations called Family Resource Centers (FRCs). The DR pilot program is administered by the Grants Management Unit within the office of the Director of the Nevada Department of Health and Human Services (DHHS) and the program has been built onto the state’s tri-level child welfare service structure. This structure includes state and county agencies and local area FRCs. DHHS, through the Division of Child and Family Services, has broad oversight responsibility for the child protection system across the state with direct responsibility for child protection services in rural Nevada. In the state’s two urban centers child protection is the primary responsibility of county agencies, the Clark County Department of Family Services and Washoe County Department of Social Services. In 1995 the state legislature established a system of Family Resource Centers to work with state and county agencies primarily to help families and individuals access needed services and support. These FRCs have been asked to play a central role in the differential response program, taking on assessment and case management functions that in other states have been handled primarily by state or county agencies. In practice, in any specific location the DR program involves the relationship between the local state or county office responsible for child welfare and the FRC responsible for the same service area. A state-level steering committee that meets regularly, and which includes representatives of the state as well as county and FRC agencies with active DR programs, provides guidance and direction to the pilot project and provides an opportunity and a mechanism for shared planning and problem solving. Staff at DHHS has organized a comprehensive training program for FRC staff related to differential response and in the use of
7
the statewide automated child welfare information system (SACWIS), which in Nevada is called UNITY.
The unique aspect of the differential response program in Nevada is the involvement of the Family Resource Centers. Reports that are selected for the differential response-family assessment by county CPS supervisors are referred to the local FRC.3
Following the initial assessment, any families that are deemed inappropriate for DR by the FRC are referred back to the county office for a formal investigation. FRC case managers are limited to a caseload of 15 DR families at any one time.
FRC case managers are then responsible for initiating contact with these families, conducting family assessments, providing ongoing services as needed, and determining when the case should be closed. The North Carolina Family Assessment Scale (NCFAS-G) is being used to assess the needs of DR families at the initial family assessment and is administered again when the case is closed. FRC staff is responsible for entering all case data on DR families into the state’s SACWIS (UNITY).
The flow chart on the following page provides a schematic diagram of the Nevada DR
model. A report of child maltreatment is received by county intake from the public or a mandated reporter, such as a physician. The report is initially screened to determine whether it meets the state’s threshold for response and should be accepted as a legitimate maltreatment report. If it does not, it is screened out. The report may nonetheless add information to an existing case (information only) or a county supervisor may decide the family could benefit from referral to another agency for assistance (information and referral) or the report may be disregarded for a variety of reasons (other).
If the report is accepted a second decision is needed: whether to refer the report to a CPS
worker for a traditional investigation or to an FRC for a DR/family assessment. Currently, this second decision is always made by a CPS supervisor.4
Families judged appropriate for a DR assessment are referred to the FRC responsible for the service area where the families live. The referrals are made with an email from CPS supervisors to FRC DR supervisors; the email includes the UNITY case number which permits the FRC to access details about the report that are in the data system.
3 In 2004, the FRC service areas were reorganized to coincide geographically with state and county child protection and other social services areas. 4 Clark County is considering modifying this procedure. The current plan is to train hotline/intake staff in what is appropriate for DR/assessment versus TR/investigation. The intake staff would then begin assigning suitable reports directly to the FRC. Reports considered in need of an investigation would be referred to a CPS supervisor who, in turn, would assign the report to a CPS field worker. A major objective of this change is to speed up report response time.
8
IntakeFRC
1. Information Only
2. Info & Referral
3. Other
Appropriate for DR Screened Out
Investigation
Appropriate for Investigation
Family Assessment
Services
Not DR Appropriate
Report
Unsubstantiated Services Out-of-Home Placement
Nevada Differential Response Flow Chart
Criteria for DR/Family Assessments. The Nevada DR pilot project is focusing on reports of child neglect, including educational, environmental, physical, medical and improper supervision. An assessment rather than a traditional CPS investigation may be offered to families with such reports unless the family has had a substantiated report in the previous three years or has had a child made a ward of the court. Families who have had three or more prior unsubstantiated reports may be referred for DR if the child welfare agency supervisors document that these reports have been reviewed before referral to an FRC. Further, state statute requires an
9
investigation for any child identified as a possible victim of abuse or neglect who is younger than 6 years.
Family Resource Centers. Currently there are five Family Resource Centers around the state that are participating in the DR pilot program. Three of the Clark County DFS offices in Las Vegas—Clark Central, Clark North, and Clark East—refer reports judged appropriate for a DR family assessment to the East Valley FRC.5
The Clark South office refers DR-appropriate reports to HopeLink FRC in Henderson. The Division of Child and Family Services (DCFS) in Elko County refers DR reports to the FRC of Northeast Nevada. And the Washoe County Department of Social Services refers prospective DR families to The Children’s Cabinet and the Washoe County FRC. The Children’s Cabinet is not technically a Family Resource Center, predating the establishment of the state FRC network and generally providing a wider set of social and community services, but for the purposes of the DR program it operates functionally much like the other FRCs.
Expansion Plans. The original plan for the implementation of the differential response approach was to eventually extend the program statewide. Currently there is no firm time table to accomplish this. It is anticipated, however, that a new round of expansion will get underway during the current state fiscal year. This is expected to include Carson City and Storey County (served by the Ron Wood Family Resource Center), Douglas County (served by the Family Support Council), Churchill County (served by the Churchill County School District/FRIENDS FRC), Lyon, Mineral and Pershing counties (served by Lyon County Human Services), southern Nye County (served by Nevada Outreach Training Organization – No to Abuse), and the Las Vegas West office in Clark County (served by Boys and Girls Club). Training will begin for the new FRC staff in November and December 2008, and DR assessments will start in each area as the staff is prepared.
Demographic Context
The differential response program has been implemented in the states two most populous counties, Clark, which includes the city of Las Vegas, and Washoe, where Reno is located, as well as in the less populous Elko County, in northeastern Nevada. Table 1 shows the population of the three current DR program counties as well as the other counties in the state along with selected demographic data that have often been found to be correlated with the relative number of child maltreatment reports. The table includes the percentage persons with children under 18 5 Initially the University Medical Center in Las Vegas provided DR services for Clark Central and Clark North offices of CCDFS but withdrew from the program and was replaced by East Valley FRC. UMC staff involved in the program moved to East Valley FRC providing service continuity.
10
in the population, the median household income, the percent of the population below the poverty level, and the percent of persons living in the same house in 2000 as in 1995, a measure of the mobility and stability of the population. The table also provides comparison figures for the state as a whole and for the country.
Table 1. Selected Population and Demographic Data
DR Program Counties
Population (2006
estimate)
Percent of state's
population
Percent of persons with
children under 18 (2006 estimate)
Median household
income (2004 dollars)
Percent of the population below the
poverty level, 2004
Living in the same house in 1995 and 2000
Clark 1,777,539 71.2% 26.0% $45,793 11.6% 34.5%
Washoe 396,428 15.9% 24.5% $50,167 10.1% 41.2%
Elko 47,114 1.9% 28.9% $52,202 8.7% 47.6%
Other Counties
Churchill 25,036 1.0% 28.0% $45,720 10.2% 45.4%
Douglas 45,909 1.8% 19.2% $54,520 7.0% 48.9%
Esmeralda 790 0.0% 18.1% $37,283 12.4% 53.1%
Eureka 1,480 0.1% 22.7% $42,790 9.0% 59.4%
Humboldt 17,446 0.7% 28.4% $47,532 9.8% 45.8%
Lander 5,272 0.2% 28.6% $49,257 9.5% 56.0%
Lincoln 4,738 0.2% 23.4% $38,226 13.0% 55.8%
Lyon 51,231 2.1% 23.5% $46,078 9.0% 44.1%
Mineral 4,868 0.2% 21.7% $33,302 14.8% 56.0%
Nye 42,693 1.7% 20.5% $41,025 11.9% 41.1%
Pershing 6,414 0.3% 22.5% $38,821 13.0% 48.4%
Storey 4,132 0.2% 18.1% $49,043 5.1% 49.1%
White Pine 9,150 0.4% 20.3% $39,420 12.4% 52.5%
Carson City 55,289 2.2% 23.1% $45,133 10.2% 46.1%
Nevada State 2,495,529 100.0% 25.4% $47,231 11.1% 37.4%
United States 299,398,484 24.6% $44,334 12.7% 54.1%
11
The state of Nevada has a somewhat higher median income than the country as a whole
as well greater mobility among its population. Within the state, the three DR pilot counties combined represent 89 percent of the state’s population. However, not all of Clark County is currently participating in the program. The four Las Vegas service areas where DR has been implemented include a little more than two-thirds of the population of Clark County. Altogether, when the population of these Clark County service areas is combined with the population of Washoe and Elko counties, the figure is close to 66 percent of the state’s total population.
The breakdown of the ethnicity of the population in the three program counties is shown
in Figure 1. The areas are predominantly white (81 percent), followed by Hispanic (26 percent), African American (9 percent) and Asian American (7 percent).
Figure 1. Ethnic Breakdown of DR Program Counties
CPS Context
Numbers of CPS Accepted Reports. Reports that are made to county offices that meet the statutory threshold for a CPS response represent the pool from which DR cases are drawn. From February 2007, when DR was first implemented in the state, through August 2008, there were 23,472 accepted reports statewide recorded in Unity. Figure 2 shows the monthly number of these reports. Three points should be made about this graph. The first is that it often takes a few months for all new child maltreatment reports to appear in state child welfare systems. Generally, then, a graph such as this will under-represent the total number of new reports during
80.5%
8.6% 6.6%1.2%
25.8%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
White Black Asian American Indian Hispanic
12
the most recent few months. The second point is that reports received by child protection systems tend to ebb and flow throughout the year, and it is frequently the case that there are fewer reports during summer months. Thirdly, however, it is nonetheless the case that the number of accepted hotline reports has decreased throughout the DR program period. This decline is not the result of the introduction of the differential response program in any way detectable to researchers, but will be monitored throughout the evaluation.
Figure 2. Monthly Number of CPS Reports Statewide Nearly two of every three maltreatment reports in the state (64 percent) from February
2007 forward have occurred in Clark County (see Figure 3.) At the same time, 20 percent have been in Washoe County, 2 percent in Elko County and 14 percent in the rest of the state. In general, these percentages reflect differences in the resident populations of these counties, although, as Figure 4 shows, there are differences between population and numbers of reports: There are somewhat fewer reports than might be expected in Clark County, for instance, and a few more in Washoe. Whether this is due to differences in the demographic characteristics of the two counties or to the initial screening process of maltreatment reports is not known presently but will be monitored.
0
200
400
600
800
1000
1200
1400
1600
1800
13
Figure 3. Percentage of Maltreatment Reports by County February 2007 – August 2008
Figure 4. Percent of the State’s Population and CPS Reports February 2007 – August 2008
Clark64%
Washoe20%
Elko2% Other
14%
63.7%
20.0%
2.3%
14.0%
71.2%
15.9%
1.9%
11.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
Clark Washoe Elko Other
CPS Reports
Population
14
Figure 5 shows the monthly number of child maltreatment reports broken down by
county. The relative population of the various counties is again evident. Also noticeable is that the decline in accepted reports was found in each of the four locations, although less in Washoe County than elsewhere.
Figure 5. Number of Maltreatment Reports by Month Statewide and by County Types of CPS Reports. CPS reports often include more than one allegation about child
maltreatment. Hotline reports received during the DR program period statewide included an average of 1.4 allegations. It is the nature of these allegations that is one factor that determines whether or not a report is suitable for a differential response or must be investigated.
Figure 6 shows how frequently different types of allegations were made in child
maltreatment reports during the program period; allegations have been collapsed into 12 categories. The most common allegation was neglect of basic needs (23.5 percent). Basic needs includes inadequate food, clothing or shelter. The second most common allegation was parental or family risks (23.1 percent), which include a variety of problems involving a child’s parents,
0
200
400
600
800
1000
1200
1400
1600
1800
Other
Elko
Washoe
Clark
total
15
such as alcohol or substance abuse, mental or physical incapacity, hospitalization or incarceration or domestic violence. Lack of proper supervision (15.6 percent) is the next most frequent allegation, followed by conflict or emotional abuse (13.2 percent) of a severe nature, and physical abuse that is considered less than severe (12.2 percent), such as excessive or inappropriate disciplining. Other allegations that are less frequently reported include sexual abuse (3.9 percent), educational neglect (2.7 percent), medical neglect (2.3 percent). Less frequent still are allegations of a drug exposed infant (1.1 percent), severe neglect (1.0 percent) such as abandonment, severe physical abuse (0.9 percent), such as poisoning and shaken baby syndrome, and unmet medical needs (0.4 percent), which generally involves a parent who is unable to access the medical care a child needs, as opposed to medical neglect, which involves a parent who fails to obtain services that are needed and available.
Figure 6. Frequency of Various Types of Allegations in Maltreatment Reports Statewide February 2007 – August 2008
Among the allegations for which current Nevada policy would permit a differential
response/family assessment are neglect of basic needs, lack of proper supervision, educational neglect, medical neglect and unmet medical needs.
Sexual abuse4%
Severe physical abuse1%
Physical abuse12%
Drug exposed infant1%
Conflict/emotional abuse13%
Severe neglect1%
Medical neglect2%
Unmet medical needs0%
Neglect of basic needs24%
Lack of proper supervision
16%
Educational neglect3%
Parental or family risks23%
16
Reports Screened for Differential Response Between the start of the DR program in February 2007 and August 2008, 563 maltreatment reports were referred to FRCs for the differential response/family assessment. Two-thirds (375) of these were in Clark County where the program was first implemented (initially in the Clark East and South service areas in February 2007 and in Clark Central and North service areas in March 2008). In Washoe County there were 165 reports referred for the differential response between January 2008, when the program was first implemented there, and August. There have been 22 referrals for differential response in Elko County from the program’s implementation there in February 2008 and August of this year. Figure 7 shows the cumulative number of DR referrals for the three counties.
Figure 7. Cumulative Number of DR/Family Assessment Referrals Since differential response was first implemented, there have been 8,479 accepted
maltreatment reports in DR program areas. This includes reports only in areas where and when a differential response was possible. The 563 reports screened appropriate for DR and referred to FRCs for a family assessment represent 6.6 percent of these reports. Figure 8 shows the percent of accepted reports screened each month for DR from the beginning of the program.
375
165
22
0
100
200
300
400
500
600
Feb0
7
Mar
07
Apr
07
May
07
Jun0
7
Jul0
7
Aug
07
Sep0
7
Oct
07
Nov
07
Dec
07
Jan0
8
Feb0
8
Mar
08
Apr
08
May
08
Jun0
8
Jul0
8
Aug
08
Elko
Washoe
Clark
17
Figure 8. Percent of Child Maltreatment Reports Screened Appropriate for DR/Assessment Figure 9 shows the percent of reports screened for DR in the three counties that have
implemented the new program. It should be noted that in Clark County the percentage includes reports in the Clark East and South service areas from February 2007 through February 2008 and
Figure 9. Percent of Reports Screened for DR/Assessment by County
2.9%
6.2%
8.1%
4.4%
9.8%
3.6%2.8%
5.1%
8.8%7.7%
10.8%
4.0%
8.6%
6.4% 6.5%
5.0%
6.5% 6.7%
12.0%
0%
2%
4%
6%
8%
10%
12%
14%
Total
0%
5%
10%
15%
20%
25%
30%
35%
40%
Clark Washoe Elko Total
18
reports from these areas plus Clark Central and North service areas from March 2008 forward. Of the three counties, Clark screened the smallest percentage for DR, 5.8 percent, followed by Washoe at 9.0 percent, and Elko, 12.1 percent.
Allegations in DR Reports. Figure 10 shows the frequency of different types of
allegations in reports that have been screened for differential response. The largest percentage (37 percent) involves families with basic needs, which nearly always indicates families in poverty. Educational neglect is also a frequent allegation in these reports (22 percent) as is lack of supervision (16 percent). Medical neglect (9 percent), parental/family risks (8 percent), conflict/emotional abuse (6 percent) and other types of allegations (2 percent) make up the rest.
Figure 10. Types of Allegations in Reports Screened for DR February 2007 – August 2008
While reports involving neglect of basic needs are the most frequent type of report
referred to FRCs for differential response, the majority of reports that included basic needs allegations were not referred for DR but were investigated by the counties. Since the beginning of the differential response program there have been 2,744 reports that included an allegation of neglect of basic needs that were screened in DR program areas. The substantial majority of these (90.4 percent) reports were investigated, while 9.6 percent were referred for DR. Some of these reports undoubtedly included other allegations that mandated an investigation, but these
Conflict/emotional abuse
6%
Neglect of basic needs37%
Educational neglect
22%
Mecial neglect
9%
Lack of supervision
16%
Parental or family risks
8%
other2%
19
represented a minority of the reports. More often such reports were excluded from a differential response because of other screening criteria—a prior substantiated report, three or more reports in the previous three years, or a victim under the age of six.
Figure 11 shows the type of response received by reports that contained allegations that
potentially would have permitted referral to an FRC for a DR. What was pointed out about reports with allegations about neglect of basic needs is true about reports with other allegations that satisfy current DR screening criteria: Some of these reports contain allegations of a more serious kind that require a formal, traditional investigation. More often, however, it is one of the other criteria for screening reports that has prevented the differential response, that is, prior CPS involvement or age of child.
Figure 11. Percent of Reports Containing Allegations that Meet DR Criteria that Received a DR/Family Assessment) or TR/Investigation
February 2007 – August 2008
Age of Children. Because of state statute that requires an investigation of a report identifying a child aged 5 or younger as a possible victim of abuse or neglect, differential response is less often utilized when maltreatment reports involve very young children. The mean age of the youngest victims in reports during the DR pilot period that were investigated was 6.3 years; for reports referred for a differential response/family assessment the mean age was 10.2 years.
21.3%9.6% 6.2%
46.0%
78.7%90.4% 93.8%
54.0%
0%10%20%30%40%50%60%70%80%90%
100%
Medical neglect/needs
Neglect of basic needs
Lack of proper supervision
Educational neglect
DR/Family Assessment TR/Investigation
20
In the 563 reports that were referred for a differential response during the initial pilot period there was an average of 1.4 children in the families, the same as for families that were investigated during this period. Figure 12 shows the percent of children by age group in families that received a DR/Family assessment or a traditional investigation. As can be seen there were few very young children in families that were referred for DR, compared to investigated families. Forty percent of the children in families that received a differential response were aged 6 to 10 years; 34 percent were teenagers. Correspondingly, traditional investigations frequently involved families with very young children: 46 percent of the children in these families were aged 5 or younger.
Figure 12. Percent of Children by Age Group in DR/Assessments and TR/Investigations This age differentiation between assessment and investigation families can be seen again in the next graph. Figure 13 is a line graph showing the age of the youngest child in families that received an investigation or a DR/family assessment. The graph plots the percent of these children by age. The graph shows that in 8.8 percent of investigations the youngest child was under 1 (<1) and that in 11.9 percent of investigations the youngest child was 1 year old. In reports referred for a differential response/family assessment the graph shows that the youngest child tended more often to be older; for example, 10.2 percent of these children were aged 7 years and 6.3 percent were 17 years old.
0.8%10.7%
3.4%
14.6%
7.3%
20.3%40.4%
26.6%14.5%
9.9%33.7%
17.9%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
DR/Assessments TR/Investigations
teens
11-12 yrs
6-10 yrs
3-5 yrs
1-2 yrs
Infants
21
Figure 13. Percent of Youngest Child in Investigations and DR/Assessments by Age
Prior CPS History. Prior contact with the child protection system is a factor when considering whether a report on a particular family is suitable for a DR/family assessment or should be investigated. A prior report of some form of child maltreatment may have been substantiated and found to be true or it may have been investigated and not substantiated.
Taken by itself, prior CPS history did not often determine whether was referred to an
FRC for a DR/assessment or whether it was investigated. Most of the time (94.8 percent) reports on families with either a prior substantiated report in the previous three years or three or more prior unsubstantiated reports were investigated. At the same time, 92.5 percent of reports on families who neither had a previous substantiated report over the last three years nor three or more unsubstantiated reports were investigated and not referred for DR/assessment.
During the DR pilot period from February 2007 through August 2008 about one in five
child maltreatment reports (21.1 percent) involved families with at least one substantiated report in the previous three years. Reports on such families were much more likely to be investigated (96.3 percent) than referred for a DR/family assessment (3.4 percent). At the same time, considering only this one factor, most reports on families with no prior substantiated reports were also more likely to be investigated (92.6 percent) than referred for a DR/assessment (7.4 percent). Looked at another way, about one in nine (11.8 percent) reports that were referred to FRCs for a DR/family assessment involved families with one or more substantiated allegations in the previous three years. Correspondingly, 21.8 percent of reports that were investigated in
10.2%
11.9%
0%
2%
4%
6%
8%
10%
12%
14%
<1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
DR Inv
22
pilot areas during this period involved families with one or more prior substantiated allegations during the last three years.
Considering prior unsubstantiated reports: 16.5 percent of child maltreatment reports in
DR pilot areas involved families with three or more prior unsubstantiated reports. Most (94.8 percent) of the reports on these families were investigated. At the same time, however, 93.1 percent of reports involved families with no unsubstantiated reports or with one or two unsubstantiated reports were likewise investigated. Considering only reports referred to FRCs for a DR/family assessment, 12.9 percent involved families with three or more prior unsubstantiated child maltreatment reports. By comparison, 16.8 percent of reports that were investigated involved families with three or more prior unsubstantiated reports. All things considered, this was not a major determining factor when considering whether or not a report would be referred for a differential response/family assessment. Combining All Factors. In Nevada, there are three primary criteria for determining suitability for a DR/assessment. First, a child maltreatment report must involve certain allegations, essentially allegations of non-severe neglect. Secondly, the report cannot involve a family with any significant CPS history. And third, reports of any sort that claim to involve child victims under the age of six must, by law, be investigated. Taken together these restrictions rule out a majority of reports. During the initial pilot period, the restrictions disallowed 83.3 percent of all reports received in areas that had implemented DR and would have permitted 16.7 percent to be referred to FRCs. Practice was even more conservative. As was seen above, the actual percent of all reports referred for a DR/assessment was 6.6 percent. Much of the difference between the possible (16.7 percent) and the actual (6.6 percent) has to do with the mechanics of setting up the complicated service structure involving Family Resource Centers and fully staffing them. A program cannot be expected to grow until it attains a steady state. Beyond that, the capacity of DR to impact the state CPS system more broadly would require policy changes that make DR eligibility criteria less restrictive. County DR Programs 1. Clark County
Currently, the differential response program is operating in four service areas within
Clark County. Three of the four areas are served by the East Valley Family Services FRC in Las Vegas. The fourth is served by HopeLink FRC in Henderson, south of Las Vegas. At the time the DR pilot began, Clark County Department of Human Services was closely focused on addressing compliance issues raised in the federally mandated Program Improvement Plan (PIP),
23
and the relationship between local CPS offices in Las Vegas and the area’s FRCs experienced a degree of stress. At the time, FRC DR workers were relatively inexperienced and needed to reconcile their service provision responsibilities with the county’s concern about child safety. Although complicated by DR staff turnover and the withdrawal of one of the original FRCs involved in the project, the county and the FRCs have been successful in working through the learning curve together.
East Valley Family Services FRC. The East Valley Family Services (EVFS) agency is
a longtime social service provider in the Las Vegas area. It currently functions as the DR provider for three Clark County service offices in the Las Vegas metropolitan area: Clark East, Clark Central and Clark North.
EVFS began providing DR services when the pilot project was first implemented in
Nevada in February 2007. At that time it was the DR arm of the Clark East office; the Clark South area was served by HopeLink FRC (see below). During the second wave of DR implementation in Clark County, in March 2008, the University Medical Center (UMC) FRC took on DR responsibilities for the county’s Las Vegas Central and North service areas. UMC’s involvement was short-lived, however, and this organization asked to withdraw from participation in the DR pilot, which occurred June 30, 2008.6
EVFS absorbed the two service areas and has been providing DR services to three Las Vegas service areas since—Clark East, Central and North.
The willingness of EVFS to take over DR services in Clark Central and North was fortuitous as the pilot had only been operating in these areas a few months. EVFS’s involvement in Clark North, however, is a temporary working arrangement. It is expected that another agency, Olive Crest,7
will take over DR services in this area by July 2009. EVFS is assisting Olive Crest to prepare for the takeover and the state is providing DR-related training. When EVFS ceases involvement in Clark North, the agency will change its name to East/Central Family Services (ECFS).
Following the decision by UMC-FRC to pull out of the DR pilot, two DR workers at UMC-FRC joined the staff at EVFS, providing continuity of services and removing an immediate need for new staff to be trained. At the present time, EVFS has four workers who conduct DR family assessments and two supervisors who also conduct family assessments. The EVFS Director has extensive experience in child welfare and is a competent child and family advocate, respected both within the FRC and the community.
6 It has been suggested that UMC withdrew at least partly because of a perceived conflict between the organization’s medical model for service provision and existing DR protocols. 7 Olive Crest was selected through an RFP process issued by the DHHS Grants Management Unit.
24
As an FRC, EVFS provides many different services in addition to those related to DR.
These other services are not directly provided by any of the DR staff (supervisors included), although the unit appears well integrated into the overall operation of the FRC. DR services are primarily provided in-house at the FRC following the family assessment and the administration of the NCFAS. When prescribed services have been provided and the family’s needs addressed in accordance with the case plan (e.g., apply for Medicaid, get housing assistance, etc.) and the case is ready to be closed, a meeting is held in the agency offices between the DR worker and family members. The case is closed in UNITY but the DR supervisor offers follow-up support to the family which, if accepted, is provided by non-DR FRC staff, generally for a period of about 30 days. A “warm hand off” is made using an in-house transfer form. The form includes background information on the family and a summary of the DR assessment. The transfer meeting is held with the family, the DR worker and the FRC case manager who will assume the case. Although the case has been closed in the information system, the case notes indicate that the family is continuing to work with an FRC worker for short-term support services. The closing summary is recorded in the case notes area of UNITY. Every contact and every attempted contact is recorded by the worker throughout the duration of the agency’s involvement with the child/family.
Services that are needed by the family but are not available directly through EVFS are
obtained through other community service providers. For example, the FRC has no ability to provide drug testing and when this is needed referrals are made to an outside source. EVFS has some FRC funds that can be utilized on a discretionary basis for meeting occasional immediate needs of children and their families. EVFS assists clients apply for assistance through the State Public Assistance system (Nomads).
EVFS staff view DR as a “perfect fit” for the work they do as an FRC. They are pleased
to be participating in the pilot project. They appreciate the steering committee meetings that have given them the sense of being involved in a broader enterprise. They also value the Clark County “Big DR” meeting involving CPS and FRC staff and administrators from DHHS and the Grants Management Unit. These sessions provide opportunities to share ideas, obtain practical assistance and resolve challenges.
HopeLink and Clark South. HopeLink (previously HACA) FRC is the agency
providing DR services in the Las Vegas South service area. The FRC is located in Henderson Nevada and was involved in the first phase of the DR pilot that began in February 2007.
HopeLink employs two DR workers. One of these workers, who generally carries only a
partial DR caseload, is also supervisor of another of the FRC’s offered programs. There have
25
been a number of personnel challenges with DR staff at HopeLink. It has been difficult for the agency to retain two fully trained DR workers for any significant period of time. At this time, the agency is on its second replacement worker since March 2008 for its first/original DR worker. With only one part-time DR worker/supervisor, and the full-time DR position occupied often by a worker in training, the agency has not often been at its full DR capacity.
This does not mean that the FRC is functioning poorly when it comes to DR, but that it
has not always been at full capacity. The skill level of the DR staff in working with clients, the attainment of a comfort level in working with often complicated child welfare cases, and the essential facility needed to use UNITY have improved at the FRC. The DR staff has increasingly recognized the importance of working closely with schools and other community service providers and their DR work has improved with this understanding. DR workers retain the case until it is closed, providing case management continuity from the first contact with the family through the last.
Communication between HopeLink and CPS South has improved considerably over
recent months, aided by bi-monthly meetings on local issues and participation in the larger statewide, steering committee meetings. Over time CPS supervisors have also gained a comfort level with DR that was not always present and tend to view the program in an increasingly positive way. Feelings that “we can work it out” together have become more common.
CPS and DR. What CPS field staff in Clark County have learned about DR has come
mostly through meetings and from the shadowing done by FRC DR workers of CPS field workers as part of DR training. CPS workers who were interviewed typically see a lot of similarity in their work and that of FRCs with DR families. CPS workers do recognize, nonetheless, that the investigations they conduct focus heavily on determining that allegations in a report are valid and that circumstances alleged did indeed occur. However, none of the CPS staff interviewed reported concern that referral for a DR family assessment placed children in greater danger. Some CPS staff noted that FRC workers have fewer tools to work with, in that CPS has access that FRC workers do not to data bases (such as police and public assistance) that can provide additional information about family member. CPS supervisors who were interviewed were generally satisfied with the criteria currently in place for screening reports for DR. Some, however, saw potential benefits from loosened restrictions or, at least, more accurately written reports. There is the perception, for example, that intake sometimes throws the blanket of victimhood over all children in a report, regardless of their involvement in an incident. If there is a child under 6 in the family who may not have been involved in an allegation but is listed as a victim, this can prevent a DR referral when an assessment might be more beneficial than an investigation. But current policies would require an
26
investigation. Sometimes the office follows up on this to ensure each child really is a victim of the allegation that was made (and is not, for example, caught up in a custody battle or harassment incident or something that does not affect all children in the family). Another example given involved reports of teenagers who were “out of control” in families “desperate for services.” If the report includes an allegation of any physical abuse it must be investigated. This may lessen the likelihood of services being provided that might have been provided if the report had been screened for DR. The result is that some families who could benefit more from a family assessment are being investigated.
2. Elko County
The Family Resource Center of Northeast Nevada (FRCNEN) is a longtime social service provider to the Elko community. The agency provides a wide variety of services, including, among others, WIC, parenting classes, early childhood development services, assorted IV-B related services, as well as DR, and helps clients access community agencies and charities for basic needs such as clothing and short-term food supplies. The agency has a new director who was hired in early 2008. The previous director remains with the agency in a grant development capacity, and provides administrative support. The new director is still trying to get her DR-sea legs firmly positioned.
The relationship between the FRC and the local CPS agency was shaken somewhat as the
DR program was getting underway when the first DR worker hired turned out to be less qualified than expected. Currently the DR staff includes one supervisor and one assessment worker, with another worker on maternity leave. The FRC’s DR staff has not been at full capacity from the start of the project in the county and this has limited referrals from the CPS office. The staff recruitment problem is attributed to the DR worker requirements established by the state, on the one hand, and the small population/low unemployment economic conditions of the area, on the other.
The DR program formally began in November 2007 in Elko, but the first family referral did not show up in the UNITY data system until February 2008. Once begun, the initial experience of adding this program to the FRC was positive. Referrals from the CPS agency were steady for a quarter, but then they stopped for a quarter because there was no DR worker available to conduct a family assessment. Referrals began again in August when a new DR worker was hired. The worker hired in August is currently on maternity leave and a new DR worker was hired in September; the potential exists for DR staffing to be at full capacity within the next quarter.
27
A second major challenge in Elko has been access to the state’s child welfare information system. Until September 2008, DR staff could not access the data system at the FRC’s offices, but had to physically go the CPS office and login there. The FRC missed a couple of critical UNITY deployments and there was no expertise in-house to move them forward. The FRC director also noted that the county’s remote location had resulted in fewer training opportunities for DR staff.
The staff of the FRC and county CPS meet twice a month. These meetings have provided opportunities for discussions about the DR program, when there was DR staff to attend them. But the relationship between the agencies extends beyond this program and CPS staff makes referrals with some frequency to the FRC for other programs as the need arises.
For their part, local CPS management expressed their full willingness to assist the FRC and their DR staff in any way they can. It is their plan to refer appropriate reports to the FRC whenever there is the DR staff to handle them. Both the county director and CPS supervisor have extensive child welfare experience and see the DR alternative as a positive development for their agency. They appear to be genuinely ready and able to support the FRC with its DR program. 3. Washoe County
The Washoe County Department of Social Services (WCDSS) is a provider of quality social services to northwest Nevada. The county is one of Nevada’s three State/DHHS defined geographic areas for child welfare and social service provision. WCDSS, in conjunction with the broader Washoe County community, acts independently to assure that funding exists for activities (including DR) that are locally seen as important, despite fiscal constraints of re-appropriation activity with the State budget for social services. This agency sees DR as the right way to deal with families in many situations (a “family friendly” approach to keeping families out of the child welfare system), and can picture DR being applied to situations far exceeding the current allowable referral criteria available in the Nevada Revised Statutes (NRS).
Organizationally, WCDSS makes DR referrals to two Washoe County service providers,
The Children’s Cabinet and the Washoe Family Resource Center. This is accomplished through the “DR Liaison” position that WCDSS internally created to meet their perceived management needs. The DR liaison functions as both the CPS supervisor for DR report assignment to the provider agencies, and as a support person for the DR workers, providing them with the necessary child welfare perspective on DR referrals and data entry support for UNITY. To date, this arrangement has worked well for all parties involved.
28
The Children’s Cabinet is not a Family Resource Center by design. It is a “home-grown” agency that predates the establishment of regional FRCs statewide and a primary social service provider for the City of Reno and Washoe County. Prior to the start of DR in the county, The Children’s Cabinet was already providing case management services for client families of WCDSS and other county agencies. The Children’s Cabinet has four DR workers and a DR supervisor on staff, in addition to many other employees. The agency has the capacity to provide a significant array of services to the local community, with numerous licensed professionals on staff and a large number of annual grants that fund ongoing operations. The first DR referrals from WCDSS to The Children’s Cabinet were made at the end of January 2008. The agency anticipates functioning at the allowable capacity for DR cases, 15 cases per DR worker. DR staff turnover has not been an issue for this agency, in comparison to some of the other DR providers in the state.
The Washoe FRC is a county-wide organization that operates out of five sites. The FRC
began as a school-based, neighborhood program in a single location. It has expanded to areas with high rates of poverty and at-risk children, working out of schools, community centers and shopping centers. It provides community-based services to meet immediate needs and increase self sufficiency, and to reduce child maltreatment through parenting education and support services. The FRC helps families access services from other community resources and provides direct and emergency services to families with basic subsistence needs, although providing fewer services directly than The Children’s Cabinet. The FRC employs two DR workers and has an FRC/DR supervisor at its location in Sparks who also works in other areas of social service delivery for the agency. Like The Children’s Cabinet, the Washoe FRC received its first DR referrals from WCDSS in January 2008.
Community support for DR exists in Washoe County. In hopes of generating greater
community awareness of DR and the community partnership, brochures were created and are being distributed locally by The Children’s Cabinet and the Washoe FRC. Within the brochure, the DR approach is fully explained and all of the partners in the process are identified and contact information provided. On April 6, 2008, the local newspaper (Daily Sparks Tribune) wrote an article that described the positive aspects of DR to the public.
Washoe County’s approach to DR and to the provision of child welfare services in
general, is unique in the state, and is one that focuses more on the needs of families than their eligibility for services. The county has been locally promoting the idea of practicing some form of differential response (as an alternative to traditional CPS practices) since the late 1990’s. In 1998, Nevada approved legislation that would permit local child welfare demonstrations related to differential response, which was viewed by the legislature as potentially keeping families off of the state’s central registry for child abuse/neglect (CA/N) reports that resulted in CPS
29
substantiation. Washoe County began to develop a differential response initiative with CA/N allegation criteria that included a broader vision than the current statewide allowable criteria now in place for DR. Washoe County’s “allowable criteria” for DR included acts of minor physical abuse that were the result of an act of discipline, and provided the programmatic assurance of traditional CPS for investigative purposes if the situation was later deemed more serious. Lack of state funding for the county-based demonstration, and lacking an outside agency with dedicated staff to provide DR assessment and subsequent services, led to less than hoped for outcomes. However, this early and innovative leaning towards DR brought the county to national attention, as differential response initiatives were the topic of discussion, and being promoted as genuine alternatives to “law and order” based CPS systems. In 2006, a publication by the American Humane Association and Child Welfare League of America, National Study on Differential Response in Child Welfare, documents Washoe County’s efforts in this work.
Historically, it has been the practice of the Washoe County community to foster their
vision of service provision locally. To that end, when the agency was advised in 2007 of the proposed state budgetary cuts that would be applied to the Children and Family Services area of governmental activity, the Washoe County consortium of community agencies and local governmental entities decided to retain their anticipated service level by financing the deficit themselves. They invested in themselves, believing the services were necessary and would yield cost-beneficial outcomes. Survey of FRC and CPS Staffs
An internet-based survey was conducted in September 2008 of FRC and CPS staff in
counties participating in Nevada’s DR pilot project. The survey was designed to capture the general attitudes, perspectives and working knowledge of FRC and county CPS staffs about the new differential response program as well as the traditional child protection system in place in the three counties. Both supervisory personnel and field workers were included in the surveys. Twelve completed surveys were submitted electronically by FRC workers and 47 by CPS workers in time to include in the present analysis. The surveys completed by the two groups were very similar; the results are summarized separately below. 1. Survey of FRC Staffs
Thirteen FRC-DR supervisors and workers responded to the survey, essentially 100
percent of the DR staff in the three counties. Twelve of the responses were received in time to be included in the analysis. Most of these individuals had been brought on relatively recently by
30
the FRCs due to the DR project. Seven of the 12 respondents had begun working at their agency in 2008, and only one reported being employed at the FRC before 2006. Ten of the respondents were DR case managers and/or conducted DR assessments, 4 had supervisory responsibilities related to DR. Current caseloads for field staff ranged from 8 to 19 families.
All FRC respondents indicated that the DR families they have worked with viewed their
agency as a source of support and assistance. Eleven of the 12 believed families felt they were better off because of their DR case. Eleven of 12 also reported feeling able to intervene in an effective way with the children and families they work with at least most of the time. All FRC staff ranked their overall ability to help families and children obtain the services and assistance they need as good to excellent.
Services. A key aspect of being able to assist parents and children effectively through
the DR program is access to services that meet family needs. Respondents were asked to review a list of possible resources and indicate what is available in their community. Table 2 shows the percent of FRC respondents who reported that specific services were provided directly by the FRC itself and/or services they assisted families receive through referrals to another agency. The table also shows the percentage of respondents who reported that a DR client was currently receiving specific services.
According to survey respondents, the services most frequently provided directly by FRCs
(according to 9 of 12 respondents) were emergency food services, assistance with utilities, rent, home repairs and other basic needs, and budgeting/financial assistance. Respondents also frequently reported that FRCs also provided assistance with transportation, employment services, parenting classes and home management services. Of the five FRCs represented by the survey, only The Children’s Cabinet respondents reported providing counseling or therapy services for children and adults.
FRCs also assist families to obtain services through referrals to other agencies. Such
referrals are frequently made for services such as childcare, medical/dental care, alcohol or drug treatment, domestic violence-related services, and legal services. The majority of respondents also indicated that other community agencies may be utilized for the same type of services that the FRCs themselves provide.
Regarding the provision of services to currently active DR families, 10 workers reported
families receiving emergency food services, 9 reported assistance with rent, utility payments or other basic needs, and 8 reported that families with active cases were receiving housing assistance, employment services and parenting training.
31
Table 2. Percent of FRC Respondents Reporting Provision and Availability of Specific Services
Services Provided by FRC
Referred to Other Agency
Current Client
Receives Services
child care 8.3% 100.0% 25.0%
respite care 33.3% 33.3% 8.3%
mental health 25.0% 91.7% 58.3%
alcohol abuse 91.7% 16.7%
other substance abuse 83.3% 16.7%
MR/DD 41.7%
medical service 91.7% 50.0%
dental services (Medicaid) 83.3% 33.3%
assistance with physical disabilities 8.3% 83.3% 50.0%
transportation 58.3% 83.3% 58.3%
DV services 8.3% 91.7% 33.3%
housing assistance 58.3% 91.7% 66.7%
assist. Utilities, rent, home repairs 75.0% 83.3% 75.0%
emergency food services/food pantry 75.0% 91.7% 83.3%
assistance with other basic needs 75.0% 83.3% 75.0%
marital or family counseling 25.0% 83.3% 58.3%
child counseling or therapy 25.0% 91.7% 58.3%
homemaker/home management assistance 58.3% 50.0% 33.3%
budgeting/financial assistance 75.0% 50.0% 50.0%
employment services 58.3% 83.3% 66.7%
adult vocational/job training 91.7% 25.0%
adult educational services 91.7% 16.7%
parenting classes 66.7% 75.0% 66.7%
household management 58.3% 33.3% 41.7%
legal services 8.3% 91.7% 50.0%
assistance with gambling addiction 8.3% 75.0%
32
FRCs and Other Agencies and Institutions. In addition to the services listed in Table
2, 11 FRC workers reported that they have frequently referred DR families to the Nevada Division of Welfare and 6 reported referrals to Job Connect. All 12 respondents describe their overall knowledge of service resources in the community as good to very good.
In general, staff at the FRCs said they had a fair-to-good working relationship with most
other major child welfare agencies and institutions in the area. They reported the FRC was on good terms with law enforcement, hospitals, and mental health facilities. An especially strong relationship was reported between the FRCs and local schools. A majority (7) of the respondents reported no working relationship between the prosecuting attorney and the FRC.
FRC View of CPS. FRC respondents were asked about their views of the child
protection system in place in their county and its ability to protect children. Many respondents were generally uncertain about this, perhaps reflecting their relatively recent involvement in child welfare. For example, 5 of the 12 respondents stated they were unsure whether the CPS system was successful in protecting children from sexual maltreatment or moderate to severe physical abuse, and 4 were unsure about the ability of CPS to protect children from lack of supervision, or medical, educational or basic needs neglect. Those respondents that expressed an opinion about CPS generally felt ambivalent to mildly positive about the system’s ability to protect children. Five respondents indicated that they believed the county CPS workers were moderately to very effective in protecting children from all forms of neglect, and 4 of the 12 believed CPS to be sufficiently effective with sexual and physical abuse.
When considering the county child protection system’s ability to effectively work with
families who have particular types of risk factors (such as substance abuse, mental illness, domestic violence, poverty, and parent-child dynamics), answers among the respondents again frequently showed uncertainty. At least four of the FRC staff marked unsure for every item they were asked about, and respondents that did express an opinion were equally split between those who believed the child protection system was effective and those that thought it was not. When asked to consider the county CPS overall, seven of the respondents reported that they were moderately to very satisfied. One reported being very dissatisfied and three indicated slight dissatisfaction.
FRC Views of DR. A majority of FRC respondents (9 of 12) described their
understanding of the goals and philosophy of the differential response approach as “thorough” and other 3 said it was “adequate.” Eleven said they were satisfied or very satisfied with the DR program in their county; one reported some dissatisfaction. A majority (9) said they needed more training related to DR.
33
To learn more about DR practice in local areas, the survey asked this question: In your
view, what are the major differences between DR and traditional CPS? The survey listed a set of practice conditions on which differences might be expected to be found. These items are listed in Table 3 along with the responses of FRC workers. Based on the DR practice model implemented in Nevada, and what is known about effective DR programs in other places, we would expect, or wish, that each of the practice conditions would be found to occur more often with a differential response than in a traditional investigation. The results seen in the table reflect a good start at best practice implementation, but also suggest a work in progress. It is an echo of the response of FRC workers that most have a clear understanding of DR and that some require more training.
Table 3. Differences Perceived by FRC Staff between DR/Assessments and Traditional CPS Investigations
DR workers
much more
likely w DR
somewhat more
likely w DR
no difference
somewhat more
likely w Inv
much more
likely w Inv
Families approached in more friendly, non-accusing manner 91.7% 8.3% 0.0% 0.0% 0.0%
No finding or substantiation of allegations 25.0% 33.3% 16.7% 8.3% 0.0%
Families more likely to receive some/any services 58.3% 16.7% 25.0% 0.0% 0.0%
Families more likely to receive services they need 50.0% 25.0% 16.7% 8.3% 0.0%
Families more likely to receive services sooner 58.3% 25.0% 8.3% 8.3% 0.0% Families more likely to be referred to other community resources 58.3% 16.7% 25.0% 0.0% 0.0%
Child less likely to be interviewed separately 16.7% 8.3% 66.7% 8.3% 0.0% More members of family tend to be present at initial assessment 25.0% 33.3% 41.7% 0.0% 0.0%
Families more cooperative 50.0% 50.0% 0.0% 0.0% 0.0% Families more likely to participate in decisions and case plans 75.0% 16.7% 8.3% 0.0% 0.0%
FRC workers were asked what aspects of DR were working well in their area and what
challenges they were experiencing. Comments about the latter included these:
34
There have been “break downs in communication between CPS and DR. We are stretching the bar when it comes to cases that are assigned to DR. At times there is a lack of professional respect in regards to DR workers’ judgment.”
There is a “lack of education about the program within other county entities: Juvenile Justice, medical institutions, welfare, schools, DFS…most of the workers have no idea about the program…”
“There are still policies and procedures to be worked out. It seems that not everyone has the same opinions on how to handle certain types of reports. The consistency within the agencies is not there.”
The FRC “workload is too demanding…”
Some FRCs have begun specific practices to strengthen their DR program. Survey respondents commented on the processes that have been working well for their office:
“We have begun to transfer closed DR cases to our home visiting FRC caseworkers—
situations where the DR process has been completed, but the family still needs additional support and/or service provision.”
There is “excellent communication between all interested parties. There is full awareness of all proposed and existing policies. Everyone knows DR can succeed, so are willing to go ‘the extra mile’ to make it succeed.”
“We maintain a close relationship with community resources. This is a huge plus in helping families.”
2. Survey of CPS Staffs Of the 47 completed surveys by county CPS professionals, 20 were submitted from Clark county, 24 from Washoe County, 3 from Elko. The job tenure of workers employed in the county CPS offices ranged from a few months to 32 years. Twenty of the respondents have been employed with CPS less than two years, and 13 have been with the agency for more than 10 years. Twelve of the CPS respondents had some supervisory responsibilities and a few of them (4) also conducted case management. The majority (62 percent) of respondents were involved in case management, many also conducting investigations. Most (87 percent) who were case managers were involved in providing services directly to families as well as referring them to community agencies for services. The average caseloads of those involved in case management and providing social services was 18.
In general, a majority of CPS workers reported a fairly positive perception of their work with families. Seven out of 10 (72.4 percent) believed they were often or always able to intervene in an effective way with children and families; 17 percent said this was the case
35
sometimes and 6.4 percent believed that they were rarely able to be effective with their interventions. The workers were asked how families tended to view CPS. About 4 in 10 (43 percent) said the families they worked with tended to view the county agency as a source of support and assistance. Another 36 percent were neutral on this question and 17 percent reported that families almost never thought of the CPS agency as a source of help. Workers were also asked whether families felt they were better off or worse off because of the involvement of the child protection agency. Just under half (47 percent) thought families tended to view themselves as somewhat-to-much better off; 34 percent were neutral in response; and 11 percent thought that families perceived themselves as worse off. Overall, workers in Washoe County were more positive in their answers to these questions, but the difference among the counties was not great.
Services. Overall, most workers felt they were able to help families obtain the services and assistance they need. Six out of 10 respondents reported that their ability to help connect families with needed services was very good. Seventeen percent perceived they had a moderate ability to help families in this way, while 15 percent did not feel positive about their abilities to help families obtain services they need.
To assist in securing appropriate services for a family, a CPS worker may refer a family
to a local social service provider. In each of the counties, this provider may be an FRC or another community agency. Respondents were asked whether there were providers for specific services in their area and they were asked whether their local FRC provided these services. Their responses are shown in Table 4.
Services which were perceived by many workers to be less accessible in the community
were respite care, MR/DD assistance, transportation, and gambling addiction treatment. According to county staff, the local FRCs were most known for providing assistance with basic needs such as utilities, rent and emergency food, as well as parenting and household management help.
Table 4 also shows the proportion of CPS staff that reported that they had referred a
client family to a particular type of service within the last month. Of the services listed, parenting classes, mental health services, substance abuse treatment and medical services were the most frequently utilized referrals, with over 50 percent of the respondents indicating that someone from a client family had recently been referred for such services. In addition to the list above, 8 out of 10 workers and supervisors also reported referring families to early childhood services and the Nevada Division of Welfare. On the whole, CPS staff viewed themselves as being knowledgeable about the service landscape in the community. Sixty-six percent of respondents believed that their knowledge of community services was good or very good, and only 15 percent viewed their knowledge as being poor or very poor.
36
Table 4. Service Providers and Services Provision Reported by CPS Staff .
Services
% of workers aware of
provider in area
% of workers
indicating FRC as provider
% of workers
referring a client in last month
Child care 87.2 14.9 46.8
Respite care 57.4 8.5 10.6
Mental health 89.4 17.0 57.4
Alcohol abuse 87.2 6.4 55.3
Other substance abuse 93.6 6.4 57.4
MR/DD 57.4 4.3 17.0
Medical services 91.5 14.9 51.1
Dental services 83.0 8.5 42.6
Assistance with physical disabilities 61.7 4.3 17.0
Transportation 59.6 21.3 44.7
DV services 87.2 12.8 48.9
Housing assistance 85.1 19.1 44.7
Assistance with utilities, rent, home repairs 78.7 40.4 38.3
Emergency food services/food pantry 87.2 38.3 48.9
Assistance with other basic needs 74.5 44.7 42.6
Marital or family counseling 83.0 19.1 44.7
Child counseling or therapy 93.6 23.4 63.8
Homemaker/home management assistance 63.8 31.9 19.1
Budgeting/financial assistance 85.1 21.3 27.7
Employment services 66.0 12.8 19.1
Adult educational services 66.0 12.8 12.8
Parenting classes 89.4 38.3 59.6
Household management 61.7 38.3 14.9
Legal services 80.9 6.4 25.5
Assistance with gambling addiction 57.4 2.1 2.1
CPS Staff Views of CPS. County CPS staff were asked to rank the ability of their own
agency to protect children from abuse and neglect. Between 45 and 55 percent of all staff believed that CPS was effective- to-very effective in protecting children against sexual abuse, physical abuse, lack of supervision, medical neglect and neglect of basic needs. Fewer respondents were strongly confident of the agency’s ability to protect children effectively from
37
educational neglect (30 percent). Other staff were more lukewarm and even pessimistic in their assessment of the agency’s effectiveness in protecting children from both severe maltreatment and child neglect, with some (4 to 9 percent) viewing the agency as generally ineffective.
CPS staff were asked further how effective they thought their agency was in working
with families with specific types of problems, including substance abuse, domestic violence, poverty, poor parenting, child behavior problems and parent-adolescent conflicts. At their most optimistic, 34 to 45 percent thought the agency was effective in working with families with substance abuse, domestic violence, and poor parenting. CPS staff were considerably more pessimistic about their agencies being able to work with families with other problems, especially extreme poverty, child behavior problems and parent-child conflicts, and families with emotional and mental health problems (where less than one in four thought their agencies were effective).
About 4 in 10 CPS workers expressed satisfaction to high satisfaction with the child
protection system in place in their county. Many were somewhat ambivalent in their response and a very small percentage (4 percent) expressed dissatisfaction. Two out of three (66 percent) expressed generally high satisfaction with their CPS job and responsibilities.
A majority (53 to 60 percent) of CPS staff reported that their agency’s relationship with
local law enforcement, juvenile court, mental health providers, hospitals and clinics was good to excellent. Other staff were more ambivalent. The relationship between CPS and the prosecuting attorney’s office and schools was nearly the same but not quite as positive. Relatively few 15-17 percent) reported that their agency had a good to excellent relationship with churches or religious organizations or with employment-related services (such as employment security or WIA).
County CPS Staff Views of DR. CPS staff were asked if they understood the goals and
philosophy of the differential response approach to child protection. A small percentage (4 percent) described their understanding as thorough and 57 percent said their understanding of DR was adequate. Of the rest, 28 percent said their understanding was less than adequate and 11 percent said it was poor. A majority said they felt the need for more training about DR; 28 percent said they needed a lot more training, 36 percent a little more. Just 17 percent said they did not feel the need for additional DR training. The rest were uncertain.
To learn more about how DR was being implemented in their areas and to gain more
complete knowledge of their understanding of this new practice, CPS workers were asked a question that was also asked of DR workers: “What are the major differences between DR and traditional CPS?” As discussed above, the survey listed a set of practice conditions on which differences might be expected to be found. As noted above, based on the Nevada DR model, if the practice were implemented as intended we would expect a majority of responses to be
38
clustered in the first two columns: these are conditions that would be expected to occur more often under differential response than a traditional investigation.
As can be seen in Table 5, a majority of CPS staff reported that DR was likely to produce
three major differences: families approached in a more friendly and non-accusing manner (63 percent), no substantiation of report (59 percent), and families more cooperative (64 percent). On other items, CPS staff expect no difference or that the condition would be more likely to occur in traditional CPS. These findings require one of two conclusions: either the DR model has not been put fully into practice or CPS workers as a group have a less than accurate understanding of DR. It is likely, as indicated above, that both explanations represent some part of the reality and that both DR implementation and CPS worker training will require on-going attention. Even though DR is the primary responsibility of FRC staff, a full understanding of the DR approach by CPS staff is important for DR to become a coherently established part of the child protection system as a whole. It is important partly because CPS staff remain ambassadors to the community and to key stakeholders in the community about DR. It cannot be expected that other stakeholders will ever have a more accurate understanding of DR than CPS workers.
Table 5. Differences Perceived by CPS Staff between DR/Assessments
and Traditional CPS Investigations
CPS workers
much more likely w DR
somewhat more
likely w DR
no difference
somewhat more likely
w/CPS
much more likely
w CPS unsure Families approached in more friendly, non-accusing manner 15.0% 47.5% 35.0% 2.5% 0.0% 14.9%
No finding or substantiation of report 34.1% 24.4% 22.0% 4.9% 14.6% 12.8% Families more likely to receive some/any services 10.0% 30.0% 55.0% 2.5% 2.5% 14.9% Families more likely to receive services they need 10.0% 32.5% 50.0% 5.0% 2.5% 14.9% Families more likely to receive services sooner 7.5% 35.0% 45.0% 10.0% 2.5% 14.9% Families more likely to be referred to other community resources 10.0% 12.5% 62.5% 12.5% 2.5% 0.0% Child less likely to be interviewed separately 25.0% 27.5% 35.0% 2.5% 10.0% 0.0% More members of family tend to be present at initial assessment 10.3% 35.9% 43.6% 7.7% 2.6% 17.0%
Families more cooperative 10.3% 53.8% 30.8% 0.0% 5.1% 17.0% Families more likely to participate in decisions and case plans 5.3% 34.2% 42.1% 13.2% 5.3% 19.1%
39
In addition, it is possible that future iterations of the Nevada child protection system will give CPS staff more direct responsibility for portions of DR. This potential eventuality would require comprehensive DR training before the fact, but would be greatly facilitated if this were a priority from the beginning.
Case-Specific Survey In order to learn more about the manner in which DR is being implemented and how it is being utilized, and to add to data available in UNITY, the case-specific survey is being conducted. Each month, as cases close, a sample is drawn and an online survey form sent to the FRC case manager. This survey is a rich source of information and has proven invaluable in similar studies of DR programs in other states. Unfortunately, as this report is being prepared only a small number of surveys have thus far been conducted. Twenty-one cases have been selected and 18 completed surveys have been received. Staff turnover has prevented completion in the other 3 cases. Currently, there are 12 additional surveys in process. The results of these will be added to those already received in the next few weeks. At the present time we can provide a summary of what has been learned from the first 18 cases, recognizing that this is very preliminary and only indicative of what is happening in DR cases and a glimpse at what can be learned through the survey.
CPS History. Of these 18 families with completed case-specific surveys, half (9) had
previous contact with child protection, half did not. One family had an open child protection case, indicating a subsequent report of consequence following the initial differential response.
Family Composition. All of the families were female headed households. The mother
of the children was present in 13 of the 18 homes (72 percent). Two of the children were in the care of their grandmothers, two were in the care of aunts and one was living with a non-related caregiver. No biological fathers were reported in the households; six adult male companions were present. Other adult males were present in three households, two grandfathers and one uncle. Five of the eighteen homes (28 percent) contained only one child under 18, eight contained two or three children, and there were five homes with four children.
FRC-Family Contacts. Face to face contact with the families from FRC workers ranged from 1 to 20 visits. In three of the cases, (Involving HopeLink, Washoe County FRC, and The Children’s Cabinet) there was one face to face visit. The one surveyed case assessed by HopeLink was found to have no legitimate allegations and the child in question was found to have no concerns. The case consisted of three telephone calls and one face to face meeting with the family, and no services were provided or recommended. In the two other cases that received
40
only one assessment visit, the worker also commented that the circumstances of the case did not require additional services; the family either refused services, or the concerns in the report were found to have resolved by the time of the visit.
Other cases reviewed in the survey were more likely to have 3 to 7 face-to-face visits. East Central Valley FRC had one case that according to the worker received 20 face to face visits and an additional 12 telephone calls. The worker identified this family as having many risk factors and needing a broad spectrum of assistance. The survey indicated that the worker addressed all of risk factors identified by assisting in securing or referring to several community services, and was successful reportedly in improving most concerns to some noticeable degree.
The number of phone contacts per case had a wider range of frequency, from 3 to 66
calls. The Washoe FRC handled a case that received 8 face-to-face visits, 66 phone contacts and 113 collateral contacts, according to the worker. Excluding the outliers, the average number of phone contacts was about eight (8.3) per family. Similarly, typical cases also received between one and six collateral contacts in which workers initiated contact with another resource or agency on behalf of the family.
Response of Families. Overall, workers found the families they approached to be fairly
cooperative during the initial visit. An exception to this was a family visited by a Washoe FRC worker that was found to be very uncooperative. This was a case that was closed after the initial assessment due to refusal of the mother to accept parenting services. East Central Valley seemed to find families most cooperative: out of four cases, three were ranked as being “very cooperative” with the worker and one moderately so. By the end of the case, workers from all FRCs found the families to be more cooperative overall.
Workers observed their families to have a range of emotional responses to their initial
assessment. The most frequently reported feelings were “pessimistic” and “satisfied.” Both of these feelings were observed by workers in 50 percent (9) of the families, but only three families apparently expressed these emotions simultaneously. Feelings of being “positive”, “worried”, and “thankful” were each observed in seven of the cases.
Risk Conditions. Multiple risk conditions were commonly found among families
reviewed by this survey. As mentioned previously, a high intensity case in East Central Valley FRC required the worker to address many risk factors, including income, employment, children school attendance, mental health, parenting, discipline, housing, physical health, and substance abuse. More typically, however, workers identified key risk factors that were a concern to the family and worked to address those particular issues.
41
The most frequently cited risk conditions were a child’s school attendance (among 61 percent of the families), the level of parenting skills in the family (44 percent), family income and employment (39 percent), a child’s school progress (33 percent), and the health of children (28 percent). In most cases case managers reported that an area identified as a risk condition was addressed while the case was open and, in a majority of cases, improvement was noted, sometimes moderate and sometimes substantial. School attendance, for example, was reported as a problem in 11 of the cases and workers reported either “much” or “moderate” improvement while the case was open. Similarly, the physical health of children also showed either moderate or much improvement in all cases in which the problem was identified. Risk factors that were identified that did not improve greatly were family income, employment and money handling skills, and the social support system for the family. These risk factors were more often cited as improving only a little.
Service Provision. Services of some kind were provided to two out of three DR
families. The remaining third did not receive any services, or direct assistance although information about services was provided to the family but the worker did not know whether or not the family followed up on it. The third that received information but no direct services included two families from East Central Valley, the single case from HopeLink, the one resistant case from the Washoe FRC, and two from The Children’s Cabinet. In East Central Valley FRC workers reported that no services were provided in one case because the child visited another state for the birth of sibling and missed school due to this event. In the other case a child missed school while a younger sibling was having major surgery, but returned to class following this incident. The Children’s Cabinet had one case where the family moved out of state two weeks after the FRC became involved. The other case involved a lice report for a child that cleared up before the FRC intervened.
Regardless of whether the presenting issue in the family was directly addressed by the
FRC’s intervention, workers tended to believe that the level of service was sufficient to reduce threats of future CA/N. In 12 of the 18 cases workers believed that the level of service was nearly or completely sufficient to avoid future risks to child safety, and 4 workers were unsure. A very similar number believed that the services they were able to provide was nearly or completely sufficient to meet other family needs as well (11 of 18, or 61 percent). Workers also indicated that the services provided were well-matched to the needs of families (11 of 18) and were either somewhat (44 percent, 8 of 18) or very effective (28 percent, 5 of 18) in solving the problems of the family.
For all cases the involvement of the extended family was not high. Seven cases were
reported to not have support from extended family at all, and only three were reported to have extensive involvement of the family. Likewise, workers reported that unfunded community
42
resources were not often used. Eleven cases were indicated to have not used this type of resource at all.
Community service providers were utilized, but not extensively, for all cases. Workers
reported connecting their families with schools (72 percent, 13 cases) and with community action agencies (4 cases) or emergency food providers (3 cases). Other types of providers, such as youth organizations, domestic violence support, or childcare providers were reportedly involved with the only one or two cases. Workers also commented that emergency gas vouchers and financial assistance were services utilized for families. According to workers, the majority of families (61 percent) did become aware of community resources that they did not know about before. In all cases workers believed that the contact with the FRC helped the family improve their ability to access needed services in the future.
Services that were reported to be provided during the case to more than one family were
educational services (5 cases), transportation (4 cases), medical or dental (3 cases) and help with basic needs (2 cases). Referrals were given in several more cases, and included parenting services (7 cases), childcare (5 cases), marital/family counseling (4 cases), and mental health (3 cases). When the service or referrals was given, parents were seen as participating most actively with the assistance with basic needs, benefit programs, education, and medical or dental care.
Workers stated that some services that were needed were not provided due to
circumstances in the case. As mentioned previously, in a few cases the child moved or the family refused services, resulting in the family being unable to complete their case plan. In another case, transportation assistance was not practical as there was no bus route in the city.
Family Surveys
Families and children are the ones most affected by any policy or practice change in child protection and they are a critical data source in the evaluation. Feedback through surveys is being sought from as many families as possible who experience the differential response. In June and July 2008 all families with closed DR cases were mailed a questionnaire that asked for their feedback and told they would receive $20 for their assistance. At the time the surveys were conducted there were 177 closed cases and all were surveyed. A significant majority (61 percent) of these surveys were returned as undeliverable due to a bad address. One completed questionnaire was received by the evaluator suggesting there may have been other mailed surveys that were not received by the intended families. By itself, this indicates the transient nature of many of the families served through CPS. At the same time it presented a major problem for the evaluation; feedback from these families is important. In response, a new
43
procedure was developed and implemented in September 2008 to make sure the questionnaire gets into the hands of family members. FRCs participating in the DR project have been sent copies of the family questionnaire along with reply envelopes. FRC staffs are being asked to hand a copy of the questionnaire and return envelope directly to families the last time the family is seen by an FRC worker. The FRC worker has also been asked to notify the evaluator, with material supplied, when a questionnaire has been given to each family. At the time this report was prepared, FRCs have reported the distribution of questionnaires to 61 families with closed DR cases and 12 completed questionnaires have been received.
As with the case-specific survey discussed above, results from this very small group of families is reported only as a early and preliminary indication of the reaction of some families to the differential response approach and to illustrate the type of information being sought from families.
Attitudes. One responding family was negative throughout the survey on nearly every
question asked and issue raised. The remaining (11 of the 12 families) were positive to very positive in most of their responses. For example, 11 (92 percent) said they were ‘satisfied’ or ‘very satisfied’ with the way they were treated by the social worker who met with them as well as with the help they received. All but two families said they were ‘better off’ or ‘much better off’ as a result of the DR intervention; the other two said they were ‘no different.’ Most said they were treated in a ‘very friendly manner’ (10) or a ‘friendly’ manner (1), and 1 described her treatment as ‘very unfriendly.’ Similarly, 10 said they were involved ‘a great deal’ in the decisions made about the family and children (1 said she was ‘somewhat’ involved and 1 said ‘not at all’). And, again, 11 said that the social worker listened ‘very much’ to what the family had to say and 11 thought the social worker tried to understand the family situation and needs ‘very much’ or ‘somewhat.’ None said that there were important issues that were not discussed.
Feelings. When asked to describe their feelings at the end of the first visit from the
social worker, 50 percent said ‘positive’; 42 percent said ‘thankful’; 33 percent said ‘hopeful,’ ‘helped,’ or ‘grateful’; 25 percent said ‘satisfied,’ ‘pleased,’ or ‘worried’; and while 17 percent said they were ‘stressed,’ an equal percent reported being ‘encouraged.’
Services. Families were asked about the services they received. Seven of the 12 said the
worker provided direct services to the family—including bus passes (for 3 families), clothing (2 families), food and school supplies (1 family each). Altogether five families said they received assistance with food or clothing for the family either directly from the worker or through another agency because of the DR worker. Two families reported receiving furniture or home repairs, child care, mental health services, help getting into education classes, counseling services, car repair or transportation assistance, help paying utilities or assistance accessing public assistance.
44
Other families said they received housing assistance, medical or dental care, help for a family member with a disability, respite care, help looking for a job, or referral to a parent support group. Ten of the families said they received the kind of services or assistance they needed and 9 said it was sufficient to meet their needs. Ten families also said the DR worker had given them referrals to community resources for help they needed and half said they had followed through so far. One in three families said the DR worker contacted other agencies on the family’s behalf. Only one respondent said they did not get the help they needed, which was to enroll a child in school.
A majority (9) of the family respondents said they were more able to care for their
children now than a year ago, and 8 said they were more confident in their ability to deal with the issues in their lives. Many still felt stresses in their lives; three in four described stresses related to their financial outlook, their current job or job prospects, or the overall well-being of their children. A third of the respondents said their income had decreased compared to the same time a year ago; half were working full time; one in three did not have health insurance for themselves or their children. Two of the families had changed addresses three or more times within the last year.
Again, it should be noted that this is a very, very small number of family respondents and
this summary is provided only to illustrate the type of information this survey is intended to obtain and to provide a small window into what is being reported to us. A few of the respondents included written comments, which suggest the value they see in the help they are getting from the FRCs.
“We are so thankful for (two DR staff names) help!” “For me, the worker was an excellent person. She was very friendly and
respectful.” “I’m very happy to have met (two DR workers) from the (FRC). They helped us
a lot to be better parents for our children. I’ll miss them a lot.”
Outcome Analysis Empirical outcome research on new programs focuses on their effects. What difference do the programs make? This may include whether they achieve the stated goals of the program, the effects on the organizations implementing the program, and most importantly, whether the clients (in this case children and families) are benefited more under the new program than they would have been under old approaches. Data for the outcome analysis include a variety of
45
information in UNITY, the results of the NCFAS assessments, and the results of the case-specific and family surveys. 1. Comparison Group Analyses
The best outcome analysis methods utilize a control or comparison group. Under this approach, DR families are taken to constitute an experimental or treatment group. Other similar families that did not receive DR are selected and collectively these are referred to as a comparison group. The analysis consists of comparing the treatment and comparison groups for differences in outcomes. If the outcomes of the treatment group are relatively more positive than those of the comparison group, they are likely attributable to the new program. Such comparisons speak to the benefits, if any, achieved through the introduction of DR. The evaluators originally assumed that all or nearly all families with appropriate reports of child abuse and neglect (CA/N) that fit the DR selection criteria would be referred to a Family Resource Center for a DR family assessment and possible services. It was noted, however, that reception of CA/N reports in quantities that were beyond the capacity of FRC agencies to handle could result in an excess of DR-appropriate families that were nonetheless approached and assessed in the traditional manner. This permits the selection of a contemporaneous comparison group of families. The process of selecting comparison families is ongoing and is discussed below. Reports Recurrence. A fundamental measure of success with families is whether or not they are encountered again by Child Protection Services. This is measured first by new accepted reports. In Nevada, a record is kept of reports that are accepted for information only (IO) or for information and referral to services (IR). Alternately certain reports are accepted with a disposition of investigation (Inv). Subsequently, the nature of the report is important. Was it the kind of report that could be assigned to a DR family assessment or the kind that requires a traditional assessment? If it was assigned to a traditional assessment, an investigation, were the allegations of the report substantiated? If it was substantiated did the extent of threats to the safety of the children necessitate removal of the children from the home? This implies four measures of recurrence:
a. Reports accepted for information only or information and referral b. Reports accepted for investigation c. Accepted reports assigned a traditional response d. Accepted reports assigned a traditional response that was substantiated e. New cases involving child removals
46
Each of these can be measured by tracking families in UNITY. Because differential response is family centered, measure of recurrence should also be family focused. Thus all reports will be counted that occur within the same family regardless of which child in the family is reported to be the alleged victim and which adult is alleged to be a perpetrator. Analysis. The most critical problem in comparing follow-up data on families in ongoing programs is differences in tracking time. By the conclusion of the evaluation, up to 36 months of data will be available for some of the project and comparison families. Only a few months of data will be available on families assigned toward the end of the project. Most families will fall somewhere in between. The first problem is how to compare families when different amounts of tracking data are available. In addition, in considering outcomes of this kind, it is important to determine not only whether the outcome occurs but also how long a period passed before the outcome occurred. For example, if 12 months of tracking data are available for two families, one of which has a new report after only 3 months while the other has a new report after 10 months the latter would be considered a more successful outcome than the former. Success is not simply avoiding any new child maltreatment but helping to create conditions in families that delay the emergence of new child maltreatment. The proper analysis to address both these issues is called survival analysis, which is a family of statistical techniques. Because families in the baseline group may, as a group, differ from families in the project group, it will be important to introduce statistical controls into the analyses. Survival analyses have been developed that permit the introduction of multiple covariates for this purpose. This approach to analysis will be used for baseline-project comparisons. 2. The North Carolina Family Assessment Scale The NCFAS-G (North Carolina Family Assessment Scale – General) will be used for project group families. This scale is used to assess families in eight domains:
• Environment • Parental capabilities • Family interactions • Family safety • Child well-being • Social/community life • Self-sufficiency • Family health
Assessments are completed at intake and at case closure. By collecting the instruments
(or the database scores entered by workers) it may be possible for evaluators to conduct pre- and
47
post-DR comparisons of project families. Pre-DR refers to the administration of the NCFAS-G at intake. Post-DR refers to the administration of the instrument at case closure. Although the reliability of the NCFAS-G has been tested on families in a California project, there is concern in Nevada that certain changes that occur in families may not be detected through pre-post comparisons. This could be a reliability issue, particularly if the instrument is not used consistently or if different workers are responsible for pre-DR and post-DR administration of the tool. However, it is also possible that the simple categorical ratings utilized in the NCFAS-G may not be sensitive enough to detect subtle changes that occurred in families—a validity concern. Analysis. The comparison of pre-DR (intake) and post-DR (closure) scores will not be of global scores (the NCFAS-G overall ratings) but scores on individual categorical items within domains (for example, not the overall environment ratings but ratings of items like housing stability, personal hygiene, etc.). If items are to be combined the combination will be based on analyses that demonstrate empirically underlying dimensions (such as factor analysis or scaling procedures). If we determine that NCFAS-G items can be used for such pre-post comparisons the proper statistics will be those that permit paired comparisons. In this case the comparison is the state of families and the situation of families at intake compared to case closure. 3. Preparation for Comparison Group Analyses By the early October 2008, 6278
cases were identified in the Nevada UNITY system as definite referrals to Family Resource Centers for DR services (Table 6). A small number of these cases may have been returned to the CPS offices from which they referred as inappropriate for DR. Only recently has a UNITY screen been implemented to track such returns. In the coming months the evaluators will be able to identify cases that have been returned to CPS. The organizational categories in Table 6 refer to the FRC agency as indicated in the initial UNITY report disposition.
Of the 627 cases, 61.2 percent were identified as closed in the UNITY data system. This may be an underestimate of closed cases, as evaluators surmise that not all case closings have been entered into UNITY. It is possible that closing of some cases coincides with the end of the “investigation” end date in the report and investigation tables and that case closing dates are not entered. Cases identified as closed had been open approximate two months (59.3 days) on average. As can be seen from the table, the length of cases was roughly comparable for each of
8 This number includes data from the UNITY extract received in mid-October 2008 and includes more complete data on September than was available when earlier sections of the report were compiled.
48
Table 6. Cases Known to be Referred to FRCs, Identified as Closed and Average (mean) Number of Days Case was Open
UNITY FRC Designation Referred Cases Percent
Identified in UNITY as
Closed
Average days open to close
Clark Central 21 3.3 21 68.3
Clark East 169 27.0 123 72.8
Clark North 95 15.2 60 41.0
Clark South 99 15.8 42 63.3
Elko 35 5.6 2 55.0
The Children’s Cabinet 132 21.1 87 57.3
Washoe FRC 76 12.1 49 43.8
Total 627 100.0 384 59.3
the seven FRC sites, and the differences among the mean days shown were not statistically significant.
Comparison Selection Method. The method for selecting comparison cases involves 1)
selecting all cases from the same county for which CA/N reports were received within a roughly comparable time period (plus or minus 60 days from the date of the target report for each DR case). Then, 2) this set of cases is reduced to those that potentially met the criteria for referral to DR but, in fact, were not referred to DR. The criteria include past substantiated and unsubstantiated reports, past state wards in the family, ages of children and neglect allegations, as discussed earlier in this report. 3) Subsequently, the best match is selected for each DR case based on the ages of children in each family, the presence of one or two caregivers and, counting from the year 2000 forward, the best match for the presence of past investigations, information-only, and information-and-referral dispositions. A final criterion involves 4) the best match from the potential comparison cases for the type of child neglect alleged in the DR report (for example educational neglect is usually matched with educational neglect).
Matching is a difficult procedure under the best of circumstances. Several matching
samples have been selected as the method is perfected. It is possible that the present sample will be modified in coming weeks. The comparison group discussed in the following table represents the best that has been selected to date.
49
The final comparison group will permit outcome comparisons as described. At this point, 618 of the 627 DR cases have been successfully matched with other similar cases that did not receive DR (Table 7).
Table 7. Comparison Cases handled by CPS Workers, Identified as Closed and Average (mean) Number of Days Case was Open
County Referred Cases Percent
Identified in UNITY as
Closed
Average days open
to close
Clark 429 69.4 198 60.3
Elko 26 4.2 19 52.4
Washoe 129 20.9 44 75.6
Carson City 34 5.5 15 55.8
Total 618 100.0 210 62.0 The cases are not exactly matched with DR cases by county, as can be seen by comparing to Table 6. The reason for this is that the number of potential matches with DR characteristics within the same date range was relatively smaller in Washoe County. Thus, cases from Clark and Carson City were used as matches. Even fewer of these cases showed a close date in UNITY than DR cases. Again, most of these cases were not substantiated and the investigation close date may be the appropriate date to use. Length of time for cases that were determined to be closed was comparable to DR cases. Interim Considerations 1. Capacity, System Impact and DR Policy
The Nevada differential response pilot project was initiated as part of the Program
Improvement Plan developed in response to the Child and Family Services Review conducted by the Administration for Children and Families in 2004. Interest in DR in the state, however, predates the CFSR and the PIP, as noted in the National Study on Differential Response in Child Welfare (2006). In 1999 legislation was adopted that permitted an alternative to an investigation. The alternative was a family assessment and services which would be allowed in response to maltreatment reports involving children older than 5 who were not in imminent danger of serious harm or threat of serious harm; the statute exempted non-excessive corporal punishment from actions requiring an investigation.
50
Nonetheless, it is the PIP that makes the issue of DR program capacity immediately
relevant. If DR is a relatively minor component in the state’s child protection system it will be limited in the leverage it can exert on the system as a whole. The larger the proportion of maltreatment reports that receive a differential response the larger the possibility this component can impact the system overall.
Given current policies, the maximum percentage of cases that can be expected to be
referred for a DR/family assessment is about 17 percent of all reports. This level is adequate for an initial test of the DR approach, but its potential impact on the child protection system as a whole is limited, no matter how positive the outcomes achieved by the new approach. Only by increasing the percentage of reports referred to DR can the potential impact be expanded.
One factor that appears to artificially deflate the size of the potential pool of DR/assessment families is that some children are classified in reports as victims who should not be. Prior to the introduction of DR it was not as important to distinguish whether or not one or both or all children in a family should be considered victims of reported allegations. However, failing to make this distinction, coupled with the investigation requirement for all reports with victims under the age of 6 years, reduces the number of reports eligible for DR. There is no way to know how often this is done or how much difference close adherence to the actual circumstances in the report would have. But we can be sure of its affect on the potential pool of DR families. We can more precisely see the effects of existing criteria on the pool of possible DR families. Of reports received during the initial pilot period, 42 percent involved allegations for which current policies allow a DR/assessment response; 66 percent involved families with a CPS history that would allow a DR/assessment; and 44 percent involved child victims over the age of 5 years. When these criteria are combined, however, only 17 percent of the reports permitted family assessments through differential response. One change that would be similar to what has been implemented effectively in some other states would be to permit reports with allegations involving less severe physical abuse, such as inappropriate discipline; about one in three reports during the initial pilot period were of this type. This change alone would increase the potential pool of DR assessments to 48 percent of the total. If, in addition to this change, the statute requiring an investigation for all reports of a child victim under 6 were amended, the potential DR pool would expand to about 7 reports in 10. One of the best run county CPS programs in the country (in Olmsted County Minnesota), conducts family assessments and not investigations on approximately this percentage of maltreatment reports.
51
The statute requiring an investigation for any report with a child victim under 6 undoubtedly is meant to provide a firm shield to protect the fundamental safety of very young children. However, a young child in a family who suffers from a lack of proper food, shelter or clothing, is as at least as certain, and more likely more certain, to receive the assistance needed through an assessment than an investigation.
All things considered, agency policies are easier to modify than state statutes.
Consideration could be given to removing the restriction on families with substantiated report in the last three years, or, at least, in removing the restriction on any substantiated prior report that would permit a family assessment now. A very young child with a basic need today who was found to have a basic need last year through a substantiated investigation, is as likely if not more likely to have the need addressed through an assessment than an investigation. In fact, it may be that the narrowly focused prior investigation never addressed or even discovered the underlying problems that gave rise to the neglect finding last year, increasing, as a result, the probability of the same finding today.
If the identification of child victims were more accurately recorded, the rule governing
prior CPS history modified, the restriction on less severe physical abuse lifted, and if the statute concerning children under 6 was limited to reports of imminent harm, we might realistically expect the potential DR pool to fall between 55 and 65 percent of reports. At that level, DR can have a significant and substantial impact on the child protection system in the state. 2. The Service Anomaly and System Adjustments
In Nevada there are regions in which the traditional child protection system has focused nearly exclusively on the immediate safety of children and less on providing services to families. Much CPS activity, therefore, revolves around cases in which children have been made wards of the state and placement has occurred. The introduction of DR offers the prospect of increasing services to families. Ironically, however, this service prospect primarily involves families in which the safety of children is less threatened and the family condition less problematic.
Differential response introduces a CPS component that is family-centered, broad in
scope, and service focused. But it concentrates on reports with less severe allegations, those in which the safety of children is not immediately threatened but in which their well being is nonetheless jeopardized. Reports involving more severe allegations that continue to receive traditional investigations are more likely to be approached with a narrow focus on the specific allegations. The underlying causes that have given rise to the problems within these families may receive less attention than the problems of families with less severe reports who receive a
52
DR assessment. Ironically, DR can introduce a process in which a broader scope of attention and a greater focus on services occur in response to reports of less severe maltreatment than is the case for reports of more severe maltreatment. Such a programmatic environment can result in less assistance being provided in situations in which more assistance is called for.
The well being of most children is inextricably tied to the well being of their families.
Enhancing the well being of the family, therefore, is the surest way to enhance the well being of children. Focusing on the immediate, short-term safety of children while ignoring their longer term welfare may have long term consequences on their safety. Examples given in the previous section are relevant again here: If families with very young children are excluded from the broader benefits of the differential response, their young children may be less likely to receive the assistance they need, either directly or through services provided to the entire family.
Similarly, if children in a family in which basic needs were found lacking in a previous
CPS investigation are excluded from having these needs addressed because the family is technically disqualified due to the ongoing nature of their problem, the children will suffer the consequences. Some of the families in the second group are excluded from DR simply because the program did not start sooner. Many of the problems substantiated two years ago would have been referred to an FRC for a family assessment if the DR program had existed. But, because of this and the fact that the earlier report was substantiated, they are ineligible for DR now; a programmatic Catch-22.
That DR generally places more emphasis upon services than traditional investigations, on
the one hand, and the specific eligibility criteria in place for a DR assessment, on the other, are two separate issues, but they have inter-connected consequences. The principle barrier to making adjustments in both instances, according to Rob Sawyer, director of the Olmsted County Child Protection Agency, is the assumption that traditional investigations make children safer. He maintains that, in most instances, a DR family assessment response is just as capable of protecting children and in the longer-term can better safeguard their well being. And, further, that just as DR can be expanded to encompass a greater proportion of families in need of services, the traditional investigative response can be improved and informed by the DR approach to families.
If you increase the pool of eligible DR families you will also need to increase the capacity of the system to serve them. This introduces both fiscal and personnel challenges. If there are more families eligible for services it will cost more to pay for services and it will require more case managers to handle their cases. The cost of services has been seen by other states as part of the price of investing in an approach that has been found over the long term to be
53
not only beneficial to children and their families but is more cost effective. But in the current economic environment, few people or institutions are investing in anything.
A second issue is nearly as daunting: To the extent that the DR program is successful, changes may be required in the CPS personnel structure. Given the public-private nature of the DR system in Nevada, this presents obvious difficulties. Successful outcomes on a scale of any significance may require more staff conducting family assessments and fewer staff conducting investigations. At some point, the possibility of developing DR staff capacity within the traditional CPS agencies may need to be considered. This capacity enhancement has a potential benefit beyond being able to provide DR assessments to a greater proportion of families. It also increases the likelihood that the traditional investigative process will be positively contaminated by the DR practice model and philosophy.
Conclusion
The differential response program in Nevada is being built on a unique foundation. The
child protection system in the state is less like the Louvre in Paris than the Guggenheim in Bilbao. It is not the staid, simple structure commonly found in states, but a system with its own swirls and flourishes. Like the system onto which it is being welded, the DR reform is complex in its form and component parts. It is part state and part county, part public and part private, part standardized and part distinct.
The child protection system in Nevada is not uniform across the state. What is feasible or even possible in one part of the state may not be feasible or possible in another. Nor will the rate of programmatic change be the same. The systems in place, for example, in Clark County and Washoe County are quite different. Part of this has to do with the community context within which the systems are situated. The presence of The Children’s Cabinet in Washoe County is a manifestation of this. The establishment of this agency is a result of and contributor to a shared vision to meet the needs of the community. It represents a service orientation in the county generally and in the Department of Social Services.
The reality is that the reality is not the same everywhere and this may always be the most
obvious fact as planning proceeds. The operating principle would seem to be: What can be done in one place, should be done there, and not be postponed because it cannot be done everywhere at once.
54
But, it is early days yet, and much has been accomplished already. The DR pilot has been built onto Nevada’s unique tri-level CPS structure. New expertise has been developed rapidly at the state, county and community level. The complex public-private implementation model has been put in place in the state’s most populous regions. Training has been provided to administrative, supervisory and field-level staff of state and county agencies and to community-level Family Resource Centers. Collaborative procedures have been designed and put into place. The steering committee of key representatives from each component of the tri-level structure meets regularly and is an effective instrument for reviewing policy and practice issues, addressing major challenges, guiding program implementation and modifications, and planning near and longer-term developments.
The pilot period is the state’s opportunity to develop, test and adjust a differential
response program that works in Nevada. System reform is easy to talk about but difficult to accomplish. Social systems have the natural momentum of large sailing ships, they are difficult to steer and take time to turn. Agreement on course and coordination of procedures are essential. Child welfare professionals with the most experience are often the most cynical observers of reform attempts. Many have seen efforts come and go, too often making changes that are only nominal in nature, changing what things are called but not really changing practice. Experience tells us, therefore, that reform must be approached soberly and built on a solid foundation, piece by piece. With its pilot project, Nevada has set a new course and a deliberate speed. The impact of DR will be small as long as current policies and state statutes restrict its use. But this is a test to see what happens on a limited scale. As stakeholders become more comfortable with the approach and as professionals become more proficient in the practice, adjustments can be made and the usage increased to the point where significant impact can be expected.